T1.2 One variable: frequency tables, bar charts, histograms Start ★ 0
T1.2Playlist 1, episode 2

What shape are the bills?

Today a long list becomes a picture whose shape you can read. Guess first: you will see your answer again later.

Builds on: T1.1 (Tag and Ruler)Leads to: T1.4, T1.5, T1.6, T4.1
0Tag and Ruler again

Three columns, three piles

Remember Tag and Ruler from T1.1? Each kind of column piles up in its own way. Press the button on each card.

To see a whole column at once, count how many fall in each pile. Measured amounts need ranges first.
1Counting labels

Count the cups

On Saturday Lina sold 40 cups. Her list is a mess. Sort it, then count.

A frequency table lists each label and how many times it appears. That count is its frequency.
In French: tableau des fréquences
1Counting labels

Out of how many?

14 mango cups: is that a lot? It depends on the total. Each square below is one of the 40 cups.

Relative frequency = frequency ÷ total. Percent frequency = relative frequency × 100. They always add up to 1, or 100%.
In French: fréquence relative, fréquence en pourcentage
2Pictures for labels

One bar per label

A bar chart gives each label its own bar. The gaps between bars say: these are separate boxes.

Bar chart: one bar per category, with gaps. For ordered labels, keep their natural order.
In French: diagramme en barres

Cup size: ordered labels keep their order

2Pictures for labels

A pie shows parts of one whole

Each slice is a share of Sunday’s 40 cups, and the slices add up to 100%. Now try to compare two slices.

Pie charts show shares of one whole. To compare close values, bars are easier: our eyes judge lengths better than angles.
In French: diagramme circulaire

Using the pie only: which slice is bigger?

3Amounts need classes

244 tables, almost no repeats

Lina dreams of a café with tables. She studies a real notebook: one waiter recorded every table he served for a few months.

For amounts, almost every value is different. So we first group the values into classes.
In French: regrouper en classes

Real data: total bill of each table, in US dollars (Bryant & Smith, 1995).

3Amounts need classes

Build classes in three steps

The textbook’s recipe: choose how many classes, find their width, set their limits.

Width ≈ (largest − smallest) ÷ number of classes, rounded up. A common guideline: 5 to 20 classes.
In French: classe, largeur de classe, limites de classe

Step 1. How many classes?

Step 2. How wide?

Step 3. Where does each class start and end?

✓ every bill fits✓ no overlaps✓ no gaps
3Amounts need classes

Count each class, then draw

Five classes of $10. Tap a row of the table and its bar rises.

A histogram is a bar chart of the classes. Its bars touch, because the number line has no gaps.
In French: histogramme
!The trap

Gaps or no gaps?

A bar chart and a histogram look alike. Their gaps tell them apart. Try breaking each rule.

Gaps: separate labels (bar chart). Touching: one number line cut into classes (histogram).

Flavors (labels)

Bills (amounts)

4Bin width

Same data, new picture

Change the number of classes. The bars change. The 244 dots underneath never move.

A new bin width changes the picture, not the data.
In French: largeur de classe
Instrument: Dot crowd. The dots stay put while the picture changes. Back in T1.3 to T1.6 and T8.5.
5Reading shape

Where is the tail?

Slide to change the shape. The name comes from the tail, not from the hump.

Skewed right: a long tail to the right. Skewed left: a long tail to the left. Symmetric: both sides match.
In French: asymétrique à droite, asymétrique à gauche, symétrique
tail on the left ◂ slide ▸ tail on the right
5Reading shape

How many humps?

Lina’s orders through the day: a lunch rush, and maybe an after-school rush. Slide the second rush away from lunch.

One hump: unimodal. Two humps: bimodal. Two humps often mean two different groups mixed together.
In French: unimodale, bimodale
invented data

5Reading shape

Read the real bills

The waiter’s 244 bills again, this time in classes of $5. Read the picture and answer.

Read a histogram in three moves: the tallest bar, the tail, and how much lies below a value.
6Running totals

Running totals

How many tables paid less than $30? Add the classes as you go, like stacking blocks.

Cumulative frequency = how many values are below the end of each class. The last one is always the total.
In French: fréquences cumulées
6Running totals

The ogive: read “less than”

Join the running totals, as percents, with a line. Drag to any bill and read how many tables paid less.

An ogive climbs from 0% to 100%. Pick a value on the bottom, go up to the line, read the percent below it.
In French: ogive, courbe des fréquences cumulées
7Stem-and-leaf

A histogram that keeps every value

The 19 Friday bills. Split each one: the tens digit is the stem, the ones digit is the leaf. Cents are dropped.

A stem-and-leaf display shows the shape like a histogram, and you can still read every value.
In French: diagramme tige et feuilles

Friday’s bills ($)

Stem | leaf (tens | ones)

8Which graph?

Which graph fits?

Pick a column. Follow the path to its graph.

Tag’s columns: bar chart or pie. Ruler’s columns: histogram, ogive, or stem-and-leaf.
9Practice

Practice arcade

Six short levels, a boss where you build the data, and a histogram lab for your own numbers. One question at a time.

10Remember

Six things to remember, and one trap

Sources and textbook links

Definitions and the three-step recipe for classes follow chapter 2 of the course textbook (frequency, relative and percent frequency distributions; bar and pie charts; classes, histograms, cumulative distributions and the ogive; the stem-and-leaf display). Shape words (skew, modality) follow OpenIntro Statistics, chapter 2. The 244 bills are real data; Lina’s cups and order times are invented for teaching. “In French” labels give common French statistical vocabulary, not quotations from the book.

  1. Anderson, D. R., Sweeney, D. J., Williams, T. A., Camm, J. D., & Cochran, J. J. (2015). Statistiques pour l’économie et la gestion (5e éd., traduction de la 7e édition américaine d’Essentials of Statistics for Business and Economics par C. Borsenberger). De Boeck Supérieur. Chapitre 2.
  2. Bryant, P. G., & Smith, M. (1995). Practical data analysis: Case studies in business statistics. Irwin. Dataset “tips” (244 tables), as distributed with the R package reshape2 and the seaborn-data repository: https://github.com/mwaskom/seaborn-data
  3. Cleveland, W. S., & McGill, R. (1984). Graphical perception: Theory, experimentation, and application to the development of graphical methods. Journal of the American Statistical Association, 79(387), 531–554. https://doi.org/10.1080/01621459.1984.10478080
  4. Diez, D. M., Çetinkaya-Rundel, M., & Barr, C. D. (2019). OpenIntro Statistics (4th ed.). OpenIntro. Chapter 2. https://www.openintro.org/book/os/